Online Machine Learning for Energy-Aware Multicore Real-Time Embedded Systems

In this article, we present an Online Learning Artificial Neural Network (ANN) model that is able to predict the performance of tasks in lower frequency levels and safely optimize real-time embedded systems' power saving operations. The proposed ANN model is supported by feature selection, whic...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on computers Ročník 71; číslo 2; s. 493 - 505
Hlavní autoři: Hoffmann, Jose Luis Conradi, Frohlich, Antonio Augusto
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York IEEE 01.02.2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Artificial neural networks
Bandwidth
Computer architecture
Embedded systems
Energy management
energy-aware systems
Learning theory
Machine learning
Monitoring
multi-core/single-chip multiprocessors
Multicore processing
Optimization
Predictive models
Real time
Real-time and embedded systems
Real-time systems
Runtime
scheduling and task partitioning
Task analysis
ISSN:0018-9340, 1557-9956
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In this article, we present an Online Learning Artificial Neural Network (ANN) model that is able to predict the performance of tasks in lower frequency levels and safely optimize real-time embedded systems' power saving operations. The proposed ANN model is supported by feature selection, which provides the most relevant variables to describe shared resource contention in the selected multicore architecture. The variables are used at runtime to produce a performance trace that encompasses sufficient information for the ANN model to predict the impact of a frequency change on the performance of tasks. A migration heuristic encompassing a weighted activity vector is combined with the ANN model to dynamically adjust frequencies and also to trigger task migrations among cores, enabling further optimization by solving resource contentions and balancing the load among cores. The proposed solution achieved energy-savings of 24.97 percent on average when compared to the run-to-end approach, and it did it without compromising the criticality of any single task. The overhead incurred in terms of execution time was 0.1791 percent on average. Each prediction added 15.3585<inline-formula><tex-math notation="LaTeX">\mu s</tex-math> <mml:math><mml:mrow><mml:mi>μ</mml:mi><mml:mi>s</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="conradihoffmann-ieq1-3056070.gif"/> </inline-formula> on average and each retraining cycle triggered at frequency adjustments was never larger than 100<inline-formula><tex-math notation="LaTeX">\mu s</tex-math> <mml:math><mml:mrow><mml:mi>μ</mml:mi><mml:mi>s</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="conradihoffmann-ieq2-3056070.gif"/> </inline-formula>.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0018-9340
1557-9956
DOI:10.1109/TC.2021.3056070